|P.Mean: Net-accessible resources for group sequential designs (created 2011-10-19).
News: Sign up for "The Monthly Mean," the newsletter that dares to call itself average, www.pmean.com/news.
Dear Professor Mean: I am trying to locate a good net-accessible resource for the group sequential designs, and sample size re-estimation for adaptive designs. Can you help?
By "net-accessible", I think you mean "free". For what it's worth, many of the trials that use group sequential designs and/or sample size re-estimation and/or adaptive designs costs thousands if not millions of dollars to run. So perhaps you may be penny wise and pound foolish here. Find out how much your trials cost to run and insist that a reasonable fraction of the cost of those trials be allocated to decent tools for planning efficient designs. I shudder at the thought of a million dollar clinical trial that does not have the budget for a ten thousand dollar software program.
Yet there is still something attractive about free net-accessible resources. If you ignore the extra time and energy that it takes for you to cobble something together from a chaotic quilt of web resources, then there is a certain satisfaction that you can achieve this way. Go ahead and use a few dozen hours of your life so you can avoid having to buy a hundred dollar book.
Alas, while much of the world's knowledge is being distributed for free these days on the web, group sequential designs lag behind. I list a few resources on my website:
It's a very haphazard list and very incomplete. It's just things I hear about on lists like this one or resources I stumble upon on my own. But it is free, which proves the saying that you get what you pay for. Note that some of these resources are publications that are not open source or they are books (which are almost never free).
There are several excellent commercial software products for this area, but it is worth noting that a free software package R has some nice add-on packages that can do some of these computations. I wrote briefly about this in 2005 at my old website:
If you look at the Comprehensive R Archive network you will find even more:
You should do a PubMed search, of course, as well as Google Scholar. If these fail to yield anything, try a general search on Google. Alas, most of the high ranking links in Google point to commercial sources. As much as I believe in the web, I think you'll find that you have to supplement your net-accessible resources with a book or two and some articles that you have to pay for (if you're not affiliated with a medical library).
I'm not sure that I'd spring for the price of some of these commercial packages though, unless you are running some very expensive clinical trials. The price of some of these packages is very steep. Try R first, but if R is too much to learn, then perhaps you need to spend the big money for a commercial package.
This page was written by Steve Simon and is licensed under the Creative Commons Attribution 3.0 United States License. Need more information? I have a page with general help resources. You can also browse for pages similar to this one at Incomplete pages.